ai-agent-skills

Available Skills

🏠 Home

🎯 Skills Index

⚙️ Installation

🤖 Agent Behavior

📋 Skill Standards

---

This is the canonical index of all available AI agent skills in the bioconductor/ai-agent-skills repository. Use this to discover and understand what skills are available and when to use them.

Note: All skills are located at skills/{skill-name}/SKILL.md.

Quick Start

1. Browse by category below to find a skill that matches your need
2. Read the skill description to understand when to use it
3. Invoke naturally: Describe what you need to your AI agent
   • Example: “Help me create a new skill”

Platform shortcuts: Optional shortcuts (like /skill-name or @workspace patterns) are documented in instructions/ for each platform.

Setup & Discovery

First time?

• See instructions/claude.md (Claude Code)
• See instructions/copilot.md (GitHub Copilot)

Understanding the design:

• See AGENTS.md for how skill discovery and invocation work

Table of Contents

• Meta Skills
• R/Bioconductor Package Skills
• Bioconductor How-Tos
• Metagenomics Skills
• Statistical Methods Skills
• Skill Categories & Tags
• Finding Skills by Use Case
• How Skills Are Discovered and Invoked
• Contributing New Skills
• Platform-Specific Information
• Questions?

Meta Skills

Infrastructure and workflow skills for working with this repository itself.

create-skill

Purpose: Help create a new AI agent skill through collaborative Q&A

When to use:

• Creating a new skill from scratch or from a manual workflow

Invocation:

• “Help me create a new skill”
• “Create a skill for [domain]”

Output: A complete skill with validated file and updated SKILLS.md, ready for testing and committing

Related skills: validate-skill, document-skill, check-bioconductor-skills

---

check-bioconductor-skills

Purpose: List and verify installed bioconductor skills

When to use:

• Verifying skills are correctly installed or discovering available skills

Invocation:

• “What skills are available?”
• “List bioconductor skills”

Output: List of available skills and setup status

---

validate-skill

Purpose: Validate that a skill conforms to the ai-agent-skills repository standards

When to use:

• Creating new skills, reviewing PRs, or ensuring standards compliance

Invocation:

• “Validate this skill”
• “Check if skills/[skill-name]/SKILL.md meets standards”

Output: Comprehensive validation report with pass/fail status, categorized issues, and specific fix suggestions

Related skills: create-skill, check-bioconductor-skills, document-skill

---

document-skill

Purpose: Automate documentation updates to SKILLS.md after creating or modifying a skill

When to use:

• After creating a new skill or modifying an existing skill to update SKILLS.md

Invocation:

• “Document the new skill I just created”
• “Update SKILLS.md for the validate-r-docs skill”

Output: Updated SKILLS.md with the skill entry added to all relevant sections

Related skills: create-skill, validate-skill, check-bioconductor-skills

---

R/Bioconductor Package Skills

Skills for analyzing, documenting, and developing R/Bioconductor packages following bioconductor conventions.

analyze-r-package

Purpose: Analyze R/Bioconductor package structure and characteristics

When to use:

• Understanding a package’s architecture, type, or identifying key functions

Invocation:

• “Analyze this R package”
• “What type of package is this?”

Output: Structured analysis of the package

---

create-package-instructions

Purpose: Generate comprehensive .github/instructions files for R/Bioconductor packages

When to use:

• Creating AI agent instructions for a new R package or standardizing documentation

Invocation:

• “Create .github/instructions for this package”
• “Generate AI agent instructions”

Output: Modular instruction files in .github/instructions/

Related skills: analyze-r-package, update-package-instructions

---

update-package-instructions

Purpose: Update existing .github/instructions files based on recent package changes

When to use:

• Refreshing package instructions after major architecture or pattern changes

Invocation:

• “Update the package instructions”
• “Refresh .github/instructions”

Output: Updated .github/instructions/ files

Related skills: analyze-r-package, create-package-instructions

---

improve-code-coverage

Purpose: Analyze R package code coverage using covr, classify testing gaps, and proactively write test cases.

When to use:

• Evaluating testing rigor, identifying gaps, and proactively writing test cases

Invocation:

• “Check my code coverage and help me write missing tests.”
• “Improve code coverage, focusing on edge cases and correctness.”

Output: A chat summary breaking down coverage by test category, and code blocks containing testthat cases.

Related skills: analyze-r-package, security-audit-r-package

---

security-audit-r-package

Purpose: Perform comprehensive security audit of R/Bioconductor packages

When to use:

• Auditing packages before CRAN/Bioconductor submission or reviewing for vulnerabilities

Invocation:

• “Run security audit on this R package”
• “Check this package for security vulnerabilities”

Output: Security audit report (markdown format) with findings categorized by severity.

Related skills: analyze-r-package, validate-skill, create-package-instructions

---

update-r-news

Purpose: Draft or update an R package NEWS file from git commit history, examining diffs when commit messages are vague

Location: skills/update-r-news/SKILL.md

When to use:

• Adding a NEWS entry before a Bioconductor release
• Drafting a changelog from recent commits
• Updating NEWS.md, NEWS, or NEWS.Rd after a development cycle
• Saving time when commit messages alone are insufficient to write clear NEWS entries

What happens:

• Detects the NEWS file format in use (NEWS.md, NEWS, or NEWS.Rd)
• Derives the upcoming even-numbered release version from the current devel version
• Finds the most recent Bioconductor devel-version bump commit (“bump x.y.z version to odd y…”) and retrieves all commits since then
• Extracts PR titles from merge commits when available
• Identifies vague commit messages (≤ 3 words or known filler phrases) and inspects their diffs to infer what changed
• Categorises changes using headings already present in the NEWS file
• Shows a full preview and asks for confirmation before writing to disk

Output: A drafted NEWS block shown for review; upon confirmation, the updated NEWS file on disk with the new version block prepended.

Related skills: analyze-r-package, update-package-instructions

---

Bioconductor How-Tos

Skills indexing focused, practical how-tos for common Bioconductor data analysis tasks.

bioc-howto

Purpose: A consolidated index of Bioconductor how-to skills covering genomics, sequencing, and omics data analysis tasks

When to use:

• Looking for step-by-step instructions for common Bioconductor tasks (e.g., loading BAM files, manipulating GRanges, extracting sequences)
• Needing examples of standard Bioconductor workflows

Invocation:

• “How do I do [task] in Bioconductor?”
• “Can you show me a Bioconductor how-to for [task]?”
• “Check the bioc-howto index for [task]”

Output: Links to and instructions from the relevant how-to markdown file.

---

Metagenomics Skills

Planned for future release

Skills for standard metagenomics data processing and analysis workflows.

---

Statistical Methods Skills

Planned for future release

Skills for statistical analysis patterns in microbiome and multi-omics research.

---

Skill Categories & Tags

By Category

Category	Skills	Purpose
meta	create-skill, check-bioconductor-skills, document-skill, validate-skill	Repository and workflow infrastructure
r-packages	analyze-r-package, create-package-instructions, improve-code-coverage, security-audit-r-package, update-package-instructions, update-r-news	R/Bioconductor package development
bioconductor-how-tos	bioc-howto	Bioconductor data analysis practical guides
metagenomics	(Planned)	Metagenomics data workflows
statistical-methods	(Planned)	Statistical analysis patterns

By Tag

Tag	Skills	Use Case
infrastructure	create-skill, check-bioconductor-skills, document-skill, validate-skill	Repository and workflow tasks
testing	improve-code-coverage	Software testing and coverage
validation	security-audit-r-package, validate-skill	Quality control and standards compliance
quality-control	security-audit-r-package, validate-skill	Ensuring skill quality
documentation	create-package-instructions, document-skill, update-package-instructions, update-r-news	Generating and maintaining docs
automation	document-skill, update-r-news	Automating repetitive documentation tasks
analysis	analyze-r-package	Understanding code and architecture
security	security-audit-r-package	Security audits and vulnerability detection
audit	security-audit-r-package	Comprehensive code auditing
bioconductor	analyze-r-package, create-package-instructions, improve-code-coverage, security-audit-r-package, update-package-instructions, update-r-news	Bioconductor-specific workflows
code-coverage	improve-code-coverage	Code coverage analysis
covr	improve-code-coverage	Tools wrapping the covr package
data-access	analyze-r-package (detects), create-package-instructions	Working with remote data
git	update-r-news	Git-based workflows
changelog	update-r-news	Changelog and release note generation
news	update-r-news	R package NEWS file management
data-analysis	bioc-howto	Practical guides for analyzing omics data

---

Finding Skills by Use Case

“I want to work on an R package…”

Use analyze-r-package to understand it, create-package-instructions to generate AI docs, improve-code-coverage or security-audit-r-package for quality control, and update-r-news for release notes.

“I want to create a new skill…”

Start with create-skill for guided help, then validate-skill and document-skill.

“I want to verify my setup…”

Run check-bioconductor-skills or check the instructions/ directory.

“I want to perform a Bioconductor data analysis task…”

Check the bioc-howto index for step-by-step guides and code examples.

“I want to analyze a workflow and automate it…”

Use create-skill.

“I want to ensure my skill meets standards…”

Use validate-skill iteratively until it passes.

---

How Skills Are Discovered and Invoked

For Users

1. Browse this file to understand what’s available
2. Describe your need to your AI agent in natural language
3. Agent handles the rest: Matches need, reads skill file, executes process, and outputs result

For AI Agents

1. Read SKILLS.md as the primary discovery mechanism
2. Match user intent to skill descriptions
3. Locate and read the skill file at skills/{skill-name}/SKILL.md
4. Execute using platform tools and deliver output

See AGENTS.md for the complete agent behavior standard.

---

Contributing New Skills

Want to add a new skill?

1. Brainstorm: Use create-skill
2. Create: Follow guidance to create the skill file
3. Validate: Run validate-skill
4. Document: Use document-skill to update this file
5. Test: Verify it works on your platform
6. Submit: Create a PR for review

See CONTRIBUTING.md for detailed guidelines.

---

Platform-Specific Information

Claude Code

See instructions/claude.md for setup and /skill-name shortcuts.

GitHub Copilot

See instructions/copilot.md for setup and @workspace shortcuts.

Other Agents

See instructions/ for available adapters.

---

Questions?

• How does skill invocation work? → See AGENTS.md
• How do I create a new skill? → Use create-skill or see CONTRIBUTING.md
• Where are the actual skill files? → skills/{skill-name}/SKILL.md
• Can I use skills on multiple platforms? → Yes! Skills are platform-agnostic
• How do I verify my setup? → Ask your agent: Do I have the Bioconductor skills? This should invoke check-bioconductor-skills if your setup is working. See instructions/ for agent-specific setup (e.g. instructions/copilot.md).
• I found an issue with a skill or want to suggest a new skill → File an issue at https://github.com/bioconductor/ai-agent-skills/issues

---

Last Updated: 2026-07-02 Version: 2.0.1 Maintained by: bioconductor

This site is open source. Improve this page.